After initially showing the decrease as an increase, it seems Valve have now corrected the Steam Hardware Survey results for October 2018.
Originally, it showed as a 0.05% increase even though the figure in October 2018 for Linux is 0.72% compared to September's 0.78% so it's a decrease of 0.06% for people to argue over. This is after Valve fixed a rounding error for small distributions that they reported on October 4th. Keep in mind, they may adjust this figure again like they have a few times.
As a reminder, we have a dedicated page tracking various details including the reported Linux market share on Steam.
It's worth noting, again, that Steam is always growing. As an example of this, Valve reported in April 2018 that they had 43 million daily active users. However, Valve did a presentation back at the Melbourne Games Week last month where they reported that daily active figure to now be 47 million. So, between April and October they had an increase of 4 million daily active users. They're also now up to 90 million monthly active users, which is 23 million more than last year (source).
To give you some fun data points:
- Linux daily active users around: 338,400.
- Linux monthly active users around: 648,000.
The amount of growth Valve is getting is pretty incredible really. They don't seem to have given out a new total active amount of users for a while, but considering their growth it seems like it could be quite a lot compared to the 125 million figure they gave out back in 2015.
So do keep in mind, that while the share is lower overall, the figure it actually represents in terms of actual Linux users is quite likely to be still growing but being outpaced by new Windows users on Steam at the same time. I imagine it's going to take a lot for our growth to ever outpace Windows and push the reported share higher to a point where it's notable. Hopefully when Steam Play matures it might help, but it's far too early to tell.